[PATCH 1/5] of: Add support for linking device tree blobs into vmlinux
David Daney
ddaney at caviumnetworks.com
Thu Nov 18 04:54:14 EST 2010
On 11/16/2010 10:14 PM, Dirk Brandewie wrote:
> On 11/16/2010 06:58 PM, Grant Likely wrote:
>> On Tue, Nov 16, 2010 at 7:21 PM, Dirk Brandewie
>> <dirk.brandewie at gmail.com> wrote:
>>>
>>> On 11/16/2010 04:39 PM, David Daney wrote:
>>>>
>>>> Thanks for doing this. However I have a few comments...
>>>>
>>>> On 11/16/2010 02:41 PM, dirk.brandewie at gmail.com wrote:
>>>>>
>>>>> From: Dirk Brandewie<dirk.brandewie at gmail.com>
>>>>>
>>>>> /* .data section */
>>>>> #define DATA_DATA \
>>>>> *(.data) \
>>>>> @@ -468,7 +482,8 @@
>>>>> MCOUNT_REC() \
>>>>> DEV_DISCARD(init.rodata) \
>>>>> CPU_DISCARD(init.rodata) \
>>>>> - MEM_DISCARD(init.rodata)
>>>>> + MEM_DISCARD(init.rodata) \
>>>>> + KERNEL_DTB()
>>>>>
>>>>
>>>> I thought the init.rodata was only for data used by __init things.
>>>> Although the
>>>> current linker scripts do not put it in the section that gets
>>>> recycled as
>>>> usable
>>>> memory.
>>>>
>>>> IIRC the unflattened version of the device tree has pointers to the
>>>> flattened
>>>> data. Since the device tree nodes are live for the entire kernel
>>>> lifecycle,
>>>> shouldn't the device tree blobs be in non-init memory?
>>>>
>>>
>>> The contents of the blob get copied to allocated memory during
>>> unflatten_device_tree() so the blob that is linked in is no longer
>>> needed
>>> after init.
>>
>> Have you written a patch to add this behaviour? The current code
>> doesn't. :-)
>>
>
> I misspoke, my blob gets copied to allocated memory during
> unflatten_device_tree.
> my early_init_dt_alloc_memory_arch() returns the physical address of a
> kmalloc'd
> buffer.
>
Perhaps you should take a look at unflatten_dt_node(), especially the
part where property names and values are assigned. I could be mistaken,
but it appears to me that the memory allocated by
early_init_dt_alloc_memory_arch() is not used to hold the name and value
strings. It is possible that they might be referred to in their
original location in the flattened blob.
> You would want copy the dtb that your platform is going to use to
> non-init memory.
>
... or you might want to locate the dtb somewhere where it would be
unlikely to get clobbered if someone were to arrange for the init.rodata
to be freed for reuse.
David Daney
More information about the devicetree-discuss
mailing list